Calibration of a fuzzy cellular automata model of urban dynamics in Saudi Arabia |
| |
| Authors: | Khalid Al-Ahmadi Linda See Alison Heppenstall James Hogg |
| |
| Institution: | 1. Department of GIS and Remote Sensing, Tarbiat Modares University, Tehran, Iran;2. RS & GIS Center, Shahid Beheshti University, Tehran, Iran;3. Geospatial Big Data Engineer, Monsanto, MO, United States;4. Faculty of Geodesy & Geomatics Eng., K.N.Toosi University of Technology, Tehran, Iran;2. James Hutton Institute, Craigiebuckler, Aberdeen AB15 8QH, Scotland, UK;3. Departamento de Geografía Humana, Universidad Complutense de Madrid, C/ Profesor Aranguren, s/n, 28040 Madrid, Spain;4. Cátedra Reducción del Riesgo de Desastres, Ciudades Resilientes, Facultad de Humanidades, Universidad de La Laguna, Campus de Guajara, s/n, 38071 Tenerife, Spain;1. Spatial Policy and Analysis Lab, Manchester Urban Institute, Department of Planning and Environmental Management, University of Manchester, Oxford Rd, Manchester M13 9PL, United Kingdom;2. CITTA, Department of Civil Engineering, University of Coimbra, Rua Luis Reis Santos, 3030-788 Coimbra, Portugal;3. Centre for Land Policy and Valuation, BarcelonaTech Universitat Politecnica de Catalunya, Av Diagonal, 649, 4a planta, 08028 Barcelona, Spain |
| |
| Abstract: | An urban cellular automata model has been designed, developed and tested for the city of Riyadh in Saudi Arabia as a research project. The model uses fuzzy set theory to capture the uncertainty associated with the transition rules and employs two automated methods of calibration: a genetic algorithm and simulated annealing. This paper describes the results of the calibration process for three time periods: 1987–1997, 1997–2005 and 1987–2005, which are characterised by different patterns of urban development. Nine different scenarios have been devised to capture the effect of different primary drivers to development including transport, urban agglomeration and topography and their interactions. The results showed that the genetic algorithm produces a better calibrated model than parallel simulated annealing. The model that contains all primary drivers and all interactions produced the best performing calibrated model overall. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
